Issues to retrieve a large number of failed jobs

[louisabraham]

Hi, my dashboard has become very very very slow. I'm talking about queries that last 5s but when cumulated cause waiting times of 30s.

By examining the requests I think the culprit are the /queues.json and /jobs/failed/1.json endpoints.

I must admit that I have 194681 failed jobs.

Also notice this contradictory information:

I'm willing to help! Where should I look?

[louisabraham]

I'm looking at the serialize_queues function

rq-dashboard/rq_dashboard/web.py

Line 107 in 3a3c91a

def serialize_queues(instance_number, queues):

but I cannot figure out where the failed queue is handled (it doesn't really exist).